PLoS ONE Alerts: Evidence-Based Healthcare

Feasibility of Provider-Initiated HIV Testing and Counselling of Tuberculosis Patients Under the TB Control Programme in Two Districts of South India

2009-11-19T08:00:00Z

Background

Provider-initiated HIV testing and counselling (PITC) is internationally recommended for tuberculosis (TB) patients, but the feasibility, effectiveness, and impact of this policy on the TB programme in India are unknown. We evaluated PITC of TB patients across two districts in India considered to have generalized HIV epidemics, Tiruchirappalli (population 2.5 million) and Mysore (population 2.8 million).

Methodology/Principal Findings

Starting June 2007, healthcare providers in both districts were instructed to ascertain HIV status for all TB patients, and refer those with unknown HIV status to the nearest Integrated Counselling and Testing Centre (ICTC)—often in the same facility—for counselling and voluntary HIV testing. All TB patients registered from June 2007 to March 2008 were followed prospectively. Field investigators assessed PITC practices and abstracted data from routine TB programme records and HIV counselling registers to determine the proportion of TB patients appropriately evaluated for HIV infection. Patient records were traced to determine the efficiency of referral links to HIV care and antiretroviral treatment (ART). Between July 2007 and March 2008, 5299 TB patients were registered in both study districts. Of the 4701 with unknown HIV status at the time of TB treatment initiation, 3368 (72%) were referred to an ICTC, and 3111 (66%) were newly tested for HIV. PITC implementation resulted in the ascertainment of HIV status for 3709/5299 (70%) of TB patients, and detected 200 cases with previously undiagnosed HIV infection. Overall, 468 (8.8%) of all registered TB patients were HIV-infected; 177 (37%) were documented to have also received any ART.

Conclusions

With implementation of PITC in India, HIV status was successfully ascertained for 70% of TB patients. Previously undiagnosed HIV-infection was detected in 6.4% of those TB patients newly tested, enabling referral for life-saving anti-retroviral treatment. ART uptake, however, was poor, suggesting that PITC implementation should include measures to strengthen and support ART referral, evaluation, and initiation.

Breaking the News or Fueling the Epidemic? Temporal Association between News Media Report Volume and Opioid-Related Mortality

2009-11-18T08:00:00Z

Background

Historical studies of news media have suggested an association between reporting and increased drug abuse. Period effects for substance use have been documented for different classes of legal and illicit substances, with the suspicion that media publicity may have played major roles in their emergence. Previous analyses have drawn primarily from qualitative evidence; the temporal relationship between media reporting volume and adverse health consequences has not been quantified nationally. We set out to explore whether we could find a quantitative relationship between media reports about prescription opioid abuse and overdose mortality associated with these drugs. We assessed whether increases in news media reports occurred before or after increases in overdose deaths.

Methodology/Principal Findings

Our ecological study compared a monthly time series of unintentional poisoning deaths involving short-acting prescription opioid substances, from 1999 to 2005 using multiple cause-of-death data published by the National Center for Health Statistics, to monthly counts of English-language news articles mentioning generic and branded names of prescription opioids obtained from Google News Archives from 1999 to 2005. We estimated the association between media volume and mortality rates by time-lagged regression analyses. There were 24,272 articles and 30,916 deaths involving prescription opioids during the seven-year study period. Nationally, the number of articles mentioning prescription opioids increased dramatically starting in early 2001, following prominent coverage about the nonmedical use of OxyContin. We found a significant association between news reports and deaths, with media reporting preceding fatal opioid poisonings by two to six months and explaining 88% (p<0.0001, df 78) of the variation in mortality.

Conclusions/Significance

While availability, structural, and individual predispositions are key factors influencing substance use, news reporting may enhance the popularity of psychoactive substances. Albeit ecological in nature, our finding suggests the need for further evaluation of the influence of news media on health. Reporting on prescription opioids conforms to historical patterns of news reporting on other psychoactive substances.

Quality and Reporting of Diagnostic Accuracy Studies in TB, HIV and Malaria: Evaluation Using QUADAS and STARD Standards

2009-11-13T08:00:00Z

Background

Poor methodological quality and reporting are known concerns with diagnostic accuracy studies. In 2003, the QUADAS tool and the STARD standards were published for evaluating the quality and improving the reporting of diagnostic studies, respectively. However, it is unclear whether these tools have been applied to diagnostic studies of infectious diseases. We performed a systematic review on the methodological and reporting quality of diagnostic studies in TB, malaria and HIV.

Methods

We identified diagnostic accuracy studies of commercial tests for TB, malaria and HIV through a systematic search of the literature using PubMed and EMBASE (2004–2006). Original studies that reported sensitivity and specificity data were included. Two reviewers independently extracted data on study characteristics and diagnostic accuracy, and used QUADAS and STARD to evaluate the quality of methods and reporting, respectively.

Findings

Ninety (38%) of 238 articles met inclusion criteria. All studies had design deficiencies. Study quality indicators that were met in less than 25% of the studies included adequate description of withdrawals (6%) and reference test execution (10%), absence of index test review bias (19%) and reference test review bias (24%), and report of uninterpretable results (22%). In terms of quality of reporting, 9 STARD indicators were reported in less than 25% of the studies: methods for calculation and estimates of reproducibility (0%), adverse effects of the diagnostic tests (1%), estimates of diagnostic accuracy between subgroups (10%), distribution of severity of disease/other diagnoses (11%), number of eligible patients who did not participate in the study (14%), blinding of the test readers (16%), and description of the team executing the test and management of indeterminate/outlier results (both 17%). The use of STARD was not explicitly mentioned in any study. Only 22% of 46 journals that published the studies included in this review required authors to use STARD.

Conclusion

Recently published diagnostic accuracy studies on commercial tests for TB, malaria and HIV have moderate to low quality and are poorly reported. The more frequent use of tools such as QUADAS and STARD may be necessary to improve the methodological and reporting quality of future diagnostic accuracy studies in infectious diseases.

Norwegian Physicians' Knowledge of and Opinions about Evidence-Based Medicine: Cross-Sectional Study

2009-11-13T08:00:00Z

Objective

To answer five research questions: Do Norwegian physicians know about the three important aspects of EBM? Do they use EBM methods in their clinical practice? What are their attitudes towards EBM? Has EBM in their opinion changed medical practice during the last 10 years? Do they use EBM based information sources?

Design

Cross sectional survey in 2006.

Setting

Norway.

Participants

966 doctors who responded to a questionnaire (70% response rate).

Results

In total 87% of the physicians mentioned the use of randomised clinical trials as a key aspect of EBM, while 53% of them mentioned use of clinical expertise and only 19% patients' values. 40% of the respondents reported that their practice had always been evidence-based. Many respondents experienced difficulties in using EBM principles in their clinical practice because of lack of time and difficulties in searching EBM based literature. 80% agreed that EBM helps physicians towards better practice and 52% that it improves patients' health. As reasons for changes in medical practice 86% of respondents mentioned medical progress, but only 39% EBM.

Conclusions

The results of the study indicate that Norwegian physicians have a limited knowledge of the key aspects of EBM but a positive attitude towards the concept. They had limited experience in the practice of EBM and were rather indifferent to the impact of EBM on medical practice. For solving a patient problem, physicians would rather consult a colleague than searching evidence based resources such as the Cochrane Library.

Targeted KRAS Mutation Assessment on Patient Tumor Histologic Material in Real Time Diagnostics

2009-11-04T08:00:00Z

Background

Testing for tumor specific mutations on routine formalin-fixed paraffin-embedded (FFPE) tissues may predict response to treatment in Medical Oncology and has already entered diagnostics, with KRAS mutation assessment as a paradigm. The highly sensitive real time PCR (Q-PCR) methods developed for this purpose are usually standardized under optimal template conditions. In routine diagnostics, however, suboptimal templates pose the challenge. Herein, we addressed the applicability of sequencing and two Q-PCR methods on prospectively assessed diagnostic cases for KRAS mutations.

Methodology/Principal Findings

Tumor FFPE-DNA from 135 diagnostic and 75 low-quality control samples was obtained upon macrodissection, tested for fragmentation and assessed for KRAS mutations with dideoxy-sequencing and with two Q-PCR methods (Taqman-minor-groove-binder [TMGB] probes and DxS-KRAS-IVD). Samples with relatively well preserved DNA could be accurately analyzed with sequencing, while Q-PCR methods yielded informative results even in cases with very fragmented DNA (p<0.0001) with 100% sensitivity and specificity vs each other. However, Q-PCR efficiency (Ct values) also depended on DNA-fragmentation (p<0.0001). Q-PCR methods were sensitive to detect ≤1% mutant cells, provided that samples yielded cycle thresholds (Ct) <29, but this condition was met in only 38.5% of diagnostic samples. In comparison, FFPE samples (>99%) could accurately be analyzed at a sensitivity level of 10% (external validation of TMGB results). DNA quality and tumor cell content were the main reasons for discrepant sequencing/Q-PCR results (1.5%).

Conclusions/Significance

Diagnostic targeted mutation assessment on FFPE-DNA is very efficient with Q-PCR methods in comparison to dideoxy-sequencing. However, DNA fragmentation/amplification capacity and tumor DNA content must be considered for the interpretation of Q-PCR results in order to provide accurate information for clinical decision making.

Cost-Effectiveness of a Telephone-Delivered Intervention for Physical Activity and Diet

2009-09-25T07:00:00Z

Background

Given escalating rates of chronic disease, broad-reach and cost-effective interventions to increase physical activity and improve dietary intake are needed. The cost-effectiveness of a Telephone Counselling intervention to improve physical activity and diet, targeting adults with established chronic diseases in a low socio-economic area of a major Australian city was examined.

Methodology/Principal Findings

A cost-effectiveness modelling study using data collected between February 2005 and November 2007 from a cluster-randomised trial that compared Telephone Counselling with a “Usual Care” (brief intervention) alternative. Economic outcomes were assessed using a state-transition Markov model, which predicted the progress of participants through five health states relating to physical activity and dietary improvement, for ten years after recruitment. The costs and health benefits of Telephone Counselling, Usual Care and an existing practice (Real Control) group were compared. Telephone Counselling compared to Usual Care was not cost-effective ($78,489 per quality adjusted life year gained). However, the Usual Care group did not represent existing practice and is not a useful comparator for decision making. Comparing Telephone Counselling outcomes to existing practice (Real Control), the intervention was found to be cost-effective ($29,375 per quality adjusted life year gained). Usual Care (brief intervention) compared to existing practice (Real Control) was also cost-effective ($12,153 per quality adjusted life year gained).

Conclusions/Significance

This modelling study shows that a decision to adopt a Telephone Counselling program over existing practice (Real Control) is likely to be cost-effective. Choosing the ‘Usual Care’ brief intervention over existing practice (Real Control) shows a lower cost per quality adjusted life year, but the lack of supporting evidence for efficacy or sustainability is an important consideration for decision makers. The economics of behavioural approaches to improving health must be made explicit if decision makers are to be convinced that allocating resources toward such programs is worthwhile.

Trial Registration

This paper uses data collected in a previous clinical trial registered at the Australian Clinical Trials Registry, Australian New Zealand Clinical Trials Registry: Anzcrt.org.au ACTRN012607000195459

Early In-Hospital Mortality following Trainee Doctors' First Day at Work

2009-09-23T07:00:00Z

Background

There is a commonly held assumption that early August is an unsafe period to be admitted to hospital in England, as newly qualified doctors start work in NHS hospitals on the first Wednesday of August. We investigate whether in-hospital mortality is higher in the week following the first Wednesday in August than in the previous week.

Methodology

A retrospective study in England using administrative hospital admissions data. Two retrospective cohorts of all emergency patients admitted on the last Wednesday in July and the first Wednesday in August for 2000 to 2008, each followed up for one week.

Principal Findings

The odds of death for patients admitted on the first Wednesday in August was 6% higher (OR 1.06, 95% CI 1.00 to 1.15, p = 0.05) after controlling for year, gender, age, socio-economic deprivation and co-morbidity. When subdivided into medical, surgical and neoplasm admissions, medical admissions admitted on the first Wednesday in August had an 8% (OR 1.08, 95% CI 1.01 to 1.16, p = 0.03) higher odds of death. In 2007 and 2008, when the system for junior doctors' job applications changed, patients admitted on Wednesday August 1^st had 8% higher adjusted odds of death than those admitted the previous Wednesday, but this was not statistically significant (OR 1.08, 95% CI 0.95 to 1.23, p = 0.24).

Conclusions

We found evidence that patients admitted on the first Wednesday in August have a higher early death rate in English hospitals compared with patients admitted on the previous Wednesday. This was higher for patients admitted with a medical primary diagnosis.

First Things First: Effectiveness and Scalability of a Basic Prehospital Trauma Care Program for Lay First-Responders in Kampala, Uganda

2009-09-11T07:00:00Z

Background

We previously showed that in the absence of a formal emergency system, lay people face a heavy burden of injuries in Kampala, Uganda, and we demonstrated the feasibility of a basic prehospital trauma course for lay people. This study tests the effectiveness of this course and estimates the costs and cost-effectiveness of scaling up this training.

Methods and Findings

For six months, we prospectively followed 307 trainees (police, taxi drivers, and community leaders) who completed a one-day basic prehospital trauma care program in 2008. Cross-sectional surveys and fund of knowledge tests were used to measure their frequency of skill and supply use, reasons for not providing aid, perceived utility of the course and kit, confidence in using skills, and knowledge of first-aid. We then estimated the cost-effectiveness of scaling up the program.

At six months, 188 (62%) of the trainees were followed up. Their knowledge retention remained high or increased. The mean correct score on a basic fund of knowledge test was 92%, up from 86% after initial training (n = 146 pairs, p = 0.0016). 97% of participants had used at least one skill from the course: most commonly haemorrhage control, recovery position and lifting/moving and 96% had used at least one first-aid item. Lack of knowledge was less of a barrier and trainees were significantly more confident in providing first-aid. Based on cost estimates from the World Health Organization, local injury data, and modelling from previous studies, the projected cost of scaling up this program was $0.12 per capita or $25–75 per life year saved. Key limitations of the study include small sample size, possible reporter bias, preliminary local validation of study instruments, and an indirect estimate of mortality reduction.

Conclusions

Lay first-responders effectively retained knowledge on prehospital trauma care and confidently used their first-aid skills and supplies for at least six months. The costs of scaling up this intervention to cover Kampala are very modest. This may be a cost-effective first step toward developing formal emergency services in Uganda other resource-constrained settings. Further research is needed in this critical area of trauma care in low-income countries.

Financing of U.S. Biomedical Research and New Drug Approvals across Therapeutic Areas

2009-09-11T07:00:00Z

Background

We estimated U.S. biomedical research funding across therapeutic areas, determined the association with disease burden, and evaluated new drug approvals that resulted from this investment.

Methodology/Principal Findings

We calculated funding from 1995 to 2005 and totaled Food and Drug Administration approvals in eight therapeutic areas (cardiovascular, endocrine, gastrointestinal, genitourinary, HIV/AIDS, infectious disease excluding HIV, oncology, and respiratory) primarily using public data. We then calculated correlations between funding, published estimates of disease burden, and drug approvals.

Financial support for biomedical research from 1995 to 2005 increased across all therapeutic areas between 43% and 369%. Industry was the principal funder of all areas except HIV/AIDS, infectious disease, and oncology, which were chiefly sponsored by the National Institutes of Health (NIH). Total (ρ = 0.70; P = .03) and industry funding (ρ = 0.69; P = .04) were correlated with projected disease burden in high income countries while NIH support (ρ = 0.80; P = .01) was correlated with projected disease burden globally. From 1995 to 2005 the number of new approvals was flat or declined across therapeutic areas, and over an 8-year lag period, neither total nor industry funding was correlated with future approvals.

Conclusions/Significance

Across therapeutic areas, biomedical research funding increased substantially, appears aligned with disease burden in high income countries, but is not linked to new drug approvals. The translational gap between funding and new therapies is affecting all of medicine, and remedies must include changes beyond additional financial investment.

Naturopathic Care for Anxiety: A Randomized Controlled Trial ISRCTN78958974

2009-08-31T07:00:00Z

Background

Anxiety is a serious personal health condition and represents a substantial burden to overall quality of life. Additionally anxiety disorders represent a significant cost to the health care system as well as employers through benefits coverage and days missed due to incapacity. This study sought to explore the effectiveness of naturopathic care on anxiety symptoms using a randomized trial.

Methods

Employees with moderate to severe anxiety of longer than 6 weeks duration were randomized based on age and gender to receive naturopathic care (NC) (n = 41) or standardized psychotherapy intervention (PT) (n = 40) over a period of 12 weeks. Blinding of investigators and participants during randomization and allocation was maintained. Participants in the NC group received dietary counseling, deep breathing relaxation techniques, a standard multi-vitamin, and the herbal medicine, ashwagandha (Withania somnifera) (300 mg b.i.d. standardized to 1.5% withanolides, prepared from root). The PT intervention received psychotherapy, and matched deep breathing relaxation techniques, and placebo. The primary outcome measure was the Beck Anxiety Inventory (BAI) and secondary outcome measures included the Short Form 36 (SF-36), Fatigue Symptom Inventory (FSI), and Measure Yourself Medical Outcomes Profile (MY-MOP) to measure anxiety, mental health, and quality of life respectively. Participants were blinded to the placebo-controlled intervention.

Results

Seventy-five participants (93%) were followed for 8 or more weeks on the trial. Final BAI scores decreased by 56.5% (p<0.0001) in the NC group and 30.5% (p<0.0001) in the PT group. BAI group scores were significantly decreased in the NC group compared to PT group (p = 0.003). Significant differences between groups were also observed in mental health, concentration, fatigue, social functioning, vitality, and overall quality of life with the NC group exhibiting greater clinical benefit. No serious adverse reactions were observed in either group.

Relevance

Many patients seek alternatives and/or complementary care to conventional anxiety treatments. To date, no study has evaluated the potential of a naturopathic treatment protocol to effectively treat anxiety. Knowledge of the efficacy, safety or risk of natural health products, and naturopathic treatments is important for physicians and the public in order to make informed decisions.

Interpretation

Both NC and PT led to significant improvements in patients' anxiety. Group comparison demonstrated a significant decrease in anxiety levels in the NC group over the PT group. Significant improvements in secondary quality of life measures were also observed in the NC group as compared to PT. The whole system of naturopathic care for anxiety needs to be investigated further including a closer examination of the individual components within the context of their additive effect.

Trial Registration

Controlled-Trials.com ISRCTN78958974

Music Attenuates Excessive Visual Guidance of Skilled Reaching in Advanced but Not Mild Parkinson's Disease

2009-08-31T07:00:00Z

Parkinson's disease (PD) results in movement and sensory impairments that can be reduced by familiar music. At present, it is unclear whether the beneficial effects of music are limited to lessening the bradykinesia of whole body movement or whether beneficial effects also extend to skilled movements of PD subjects. This question was addressed in the present study in which control and PD subjects were given a skilled reaching task that was performed with and without accompanying preferred musical pieces. Eye movements and limb use were monitored with biomechanical measures and limb movements were additionally assessed using a previously described movement element scoring system. Preferred musical pieces did not lessen limb and hand movement impairments as assessed with either the biomechanical measures or movement element scoring. Nevertheless, the PD patients with more severe motor symptoms as assessed by Hoehn and Yahr (HY) scores displayed enhanced visual engagement of the target and this impairment was reduced during trials performed in association with accompanying preferred musical pieces. The results are discussed in relation to the idea that preferred musical pieces, although not generally beneficial in lessening skilled reaching impairments, may normalize the balance between visual and proprioceptive guidance of skilled reaching.

Business Return in New Orleans: Decision Making Amid Post-Katrina Uncertainty

2009-08-26T07:00:00Z

Background

Empirical observations on how businesses respond after a major catastrophe are rare, especially for a catastrophe as great as Hurricane Katrina, which hit New Orleans, Louisiana on August 29, 2005. We analyzed repeated telephone surveys of New Orleans businesses conducted in December 2005, June 2006, and October 2007 to understand factors that influenced decisions to re-open amid post-disaster uncertainty.

Methodology/Principal Findings

Businesses in the group of professional, scientific, and technical services reopened the fastest in the near term, but differences in the rate of reopening for businesses stratified by type became indistinguishable in the longer term (around two years later). A reopening rate of 65% was found for all businesses by October 2007. Discriminant analysis showed significant differences in responses reflecting their attitudes about important factors between businesses that reopened and those that did not. Businesses that remained closed at the time of our third survey (two years after Katrina) ranked levee protection as the top concern immediately after Katrina, but damage to their premises and financing became major concerns in subsequent months reflected in the later surveys. For businesses that had opened (at the time of our third survey), infrastructure protection including levee, utility, and communications were the main concerns mentioned in surveys up to the third survey, when the issue of crime became their top concern.

Conclusions/Significance

These findings underscore the need to have public policy and emergency plans in place prior to the actual disaster, such as infrastructure protection, so that the policy can be applied in a timely manner before business decisions to return or close are made. Our survey results, which include responses from both open and closed businesses, overcome the “survivorship bias” problem and provide empirical observations that should be useful to improve micro-level spatial economic modeling of factors that influence business return decisions.

Evaluation of Interferon-Gamma Release Assays in the Diagnosis of Recent Tuberculosis Infection in Health Care Workers

2009-08-24T07:00:00Z

Background

Health care workers (HCWs) are a group at risk of latent tuberculosis infection (LTBI). The aims of this study were to determine IFN-γ response by QuantiFERON-TB GOLD In Tube (QFN-G-IT) and T-SPOT.TB in HCWs, comparing the results with tuberculin skin test (TST); and to analyze the capacity of IFN-γ tests to detect recent versus remote LTBI with a prolonged stimulation test (PST).

Methodology/Principal Findings

A total of 147 HCWs were enrolled; 23 of whom were BCG vaccinated. 95 HCWs (64.6%) had a previous positive TST and were not retested; and 52 HCWs had a previous negative TST or were tested for the first time. When we analysed individuals without previous positive TST, the number of positive results for T-SPOT.TB was 12/52 (23.1%); and for QFN-G-IT, 9/52 (17.3%). The global concordance (κ) between T-SPOT.TB and QFN-G-IT with TST was 0.754 and 0.929 respectively. Of individuals with previous positive TST, T-SPOT.TB and QFN-G-IT were negative in 51.6% (49/95) and 62.1% (59/95) respectively, decreasing the concordance to 0.321 and 0.288, respectively. In non-BCG vaccinated HCWs with previous positive TST a positive IFN-γ test was associated with degree of exposure and diameter of TST. PST was performed in 24 HCW with previous positive TST and negative IFN-γ tests. PST was developed in 3 cell cultures stimulated with medium alone, ESAT-6 and CFP-10, respectively. In the third and sixth day of incubation period, part of the supernatants were replaced with complete medium supplemented with (rIL)-2. On day 9, ELISPOT assay was performed. In 14 samples PST was not valid due to not having enough cells. In 8 cases, the response was negative, and in 2 cases positive, suggesting that these patients were infected with Mycobacterium tuberculosis in some point in the past.

Conclusions

Both IFN-γ tests showed a similar number of positive results, and concordance between the tests was excellent. None of the tests was affected by prior BCG vaccination. IFN-γ tests are a useful tool for detecting recent infection in HCW population.

Estimating Sensitivity of Laboratory Testing for Influenza in Canada through Modelling

2009-08-18T07:00:00Z

Background

The weekly proportion of laboratory tests that are positive for influenza is used in public health surveillance systems to identify periods of influenza activity. We aimed to estimate the sensitivity of influenza testing in Canada based on results of a national respiratory virus surveillance system.

Methods and Findings

The weekly number of influenza-negative tests from 1999 to 2006 was modelled as a function of laboratory-confirmed positive tests for influenza, respiratory syncytial virus (RSV), adenovirus and parainfluenza viruses, seasonality, and trend using Poisson regression. Sensitivity was calculated as the number of influenza positive tests divided by the number of influenza positive tests plus the model-estimated number of false negative tests. The sensitivity of influenza testing was estimated to be 33% (95%CI 32–34%), varying from 30–40% depending on the season and region.

Conclusions

The estimated sensitivity of influenza tests reported to this national laboratory surveillance system is considerably less than reported test characteristics for most laboratory tests. A number of factors may explain this difference, including sample quality and specimen procurement issues as well as test characteristics. Improved diagnosis would permit better estimation of the burden of influenza.

Missing Data in Randomized Clinical Trials for Weight Loss: Scope of the Problem, State of the Field, and Performance of Statistical Methods

2009-08-13T07:00:00Z

Background

Dropouts and missing data are nearly-ubiquitous in obesity randomized controlled trails, threatening validity and generalizability of conclusions. Herein, we meta-analytically evaluate the extent of missing data, the frequency with which various analytic methods are employed to accommodate dropouts, and the performance of multiple statistical methods.

Methodology/Principal Findings

We searched PubMed and Cochrane databases (2000–2006) for articles published in English and manually searched bibliographic references. Articles of pharmaceutical randomized controlled trials with weight loss or weight gain prevention as major endpoints were included. Two authors independently reviewed each publication for inclusion. 121 articles met the inclusion criteria. Two authors independently extracted treatment, sample size, drop-out rates, study duration, and statistical method used to handle missing data from all articles and resolved disagreements by consensus. In the meta-analysis, drop-out rates were substantial with the survival (non-dropout) rates being approximated by an exponential decay curve (e^−λt) where λ was estimated to be .0088 (95% bootstrap confidence interval: .0076 to .0100) and t represents time in weeks. The estimated drop-out rate at 1 year was 37%. Most studies used last observation carried forward as the primary analytic method to handle missing data. We also obtained 12 raw obesity randomized controlled trial datasets for empirical analyses. Analyses of raw randomized controlled trial data suggested that both mixed models and multiple imputation performed well, but that multiple imputation may be more robust when missing data are extensive.

Conclusion/Significance

Our analysis offers an equation for predictions of dropout rates useful for future study planning. Our raw data analyses suggests that multiple imputation is better than other methods for handling missing data in obesity randomized controlled trials, followed closely by mixed models. We suggest these methods supplant last observation carried forward as the primary method of analysis.